Evaluating establishment success of non-native fishes introduced to inland aquatic habitats of tropical Pacific islands

Page created by John Graham
 
CONTINUE READING
Evaluating establishment success of non-native fishes introduced to inland aquatic habitats of tropical Pacific islands
Journal of
                                                                                                                  Open Access
       Vertebrate Biology
       J. Vertebr. Biol. 2021, 70(4): 21064                                                        DOI: 10.25225/jvb.21064
                                              SPECIAL ISSUE: INVASIVE FISHES
                                                    RESEARCH PAPER

        Evaluating establishment success of non-native
         fishes introduced to inland aquatic habitats of
                     tropical Pacific islands
                           Stephen J. WALSH1*, Leo G. NICO1 and Mark W. MILLER2

1
    U.S. Geological Survey, Wetland and Aquatic Research Center, Gainesville, Florida, USA; e-mail: swalsh@usgs.gov,
    lnico@usgs.gov
2
    Integrated Statistics, Woods Hole, Massachusetts, USA; e-mail: mwmilleraws@gmail.com

Received 26 August 2021; Accepted 19 October 2021; Published online 9 December 2021

Abstract. An information-theoretic approach was used to evaluate non-native freshwater fish species
introduced to insular habitats of Hawaii and Guam comparing successful establishments vs. failures. Since
the late 1800s, as many as 81 non-native freshwater fish species have been recorded as introduced to Hawaii
and Guam (combined) and 50 (62%) of these are documented as having one or more established populations.
We examined eleven independent variables to investigate establishment success by creating 21 a priori logistic
regression models ranked using Akaike’s Information Criterion adjusted for small sample size. An additional
eight post-hoc models were included that comprised the best a priori model and various combinations of
individual variables. The best overall model of establishment probability included effects of taxonomic
affinity (family membership), prior establishment success on other tropical islands, and hypoxia tolerance.
Establishment success in Hawaii and Guam was highest for those species established on many other islands,
and according to our best model air-breathing fishes were more likely to become established. Six fish families,
each with from three to 18 species introduced to Pacific islands, were highly successful at establishment:
Cichlidae (16 established of 18 species introduced), Poeciliidae (seven of eight), Cyprinidae (four of seven),
Centrarchidae (four of four), Clariidae (three of three), and Loricariidae (three of four). Those that successfully
established include both small and moderately large-bodied taxa, while representing a diverse array of other
morphological and life-history traits. Pathways and motives associated with fish introductions in the Pacific
have been linked to desires to develop aquaculture, enhance wild stocks of food, sport, and bait fishes, for
use as biological control agents, or are linked to the ornamental fish trade. We found that many established
species were introduced via multiple pathways (up to eight) and our analyses suggest that the combination of
prior establishment success on other tropical islands and presence of non-native fishes in multiple pathways
was indicative of high propagule pressure. Our study results and conclusions on Pacific tropical island
introductions are in general agreement with previous studies on non-native freshwater fishes in other regions
of the world and similar to observations in continental ecosystems and temperate zones.

Key words: nonindigenous fishes, Hawaii, Guam, Micronesia, prior establishment success, hypoxia tolerance,
air breathing

                                                                This is an open access article under the terms of the Creative Commons
                                                                Attribution Licence (CC BY 4.0), which permits use, distribution and
* Corresponding Author                                          reproduction in any medium provided the original work is properly cited.
Evaluating establishment success of non-native fishes introduced to inland aquatic habitats of tropical Pacific islands
J. Vertebr. Biol. 2021, 70(4): 21064                     2               Inland fish introductions on tropical Pacific islands

Introduction                                             ecological generalists, behaviourally flexible,
                                                         highly tolerant to abiotic stressors, capable of rapid
A major focus of invasion biology is the screening       dispersal, exploitative of disturbed habitats, more
and profiling of species for the purpose of              often classified as r-selected rather than k-selected
identifying non-native taxa predisposed to               and, affiliated in some way to humans (Ehrlich
becoming successful invaders (Kolar 2004,                1986, 1989, Lodge 1993, Ricciardi & Rasmussen
Marchetti et al. 2004a, b, Howeth et al. 2016,           1998, Kolar & Lodge 2002, Alcaraz et al. 2005).
Meyers et al. 2020). The process typically involves      Admittedly, many studies have found that some or
gathering information on a species’ biological           many of the above listed traits and descriptors are
attributes and determining which traits, if any,         not necessarily the best predictors of establishment
are advantageous to establishment in a novel             or invasion success, and that differences may
ecosystem. Such assessments require detailed             depend on the phylogenetic group, the geographic
knowledge of the ecology and life history of an          region, and the invasion process stage examined
organism in its native range, although information       (Kolar & Lodge 2001, Ruesink 2005, García-Berthou
about a species’ introduction history is equally         2007, Gozlan et al. 2010).
important. Mechanisms or conditions that promote
the invasion process, such as existence of transport     To better understand the complex relationship
pathways, climate match, and propagule pressure          between invasion success and species characteristics,
are also commonly part of many such analyses.            investigators increasingly apply quantitative
Consequently, predicting the identity or range of        modelling approaches to the problem (Kolar &
potential invaders commonly involves selection,          Lodge 2001, 2002, Kolar 2004, Marchetti et al.
categorization, analysis, and assessment of an           2004a, Moyle & Marchetti 2006, Howeth et al. 2016).
array of variables (Andersen et al. 2004, Copp et al.    Comparisons of the results of studies, whether
2009, Keller et al. 2011, Howeth et al. 2016, Lodge      quantitative or qualitative, is challenging because
et al. 2016, Davidson et al. 2017).                      of differences among studies in the variables
                                                         selected for examination, the organisms and
Terms and concepts in invasion biology are not           spatial scales evaluated, and the specific analytical,
widely standardized; consequently, there exist           statistical, and interpretive methods used (García-
nuanced and, in some cases, marked differences in        Berthou 2007). Factors associated with failed
the way different authors have defined or used them      introductions may be equally as important as those
(Fuller et al. 1999, Davis & Thompson 2000, Colautti     leading to successful establishment. However, there
& MacIsaac 2004, Copp et al. 2005a, b). Herein, the      are often few data available on failed introductions
term “establishment” refers to a non-native species      and suggested reasons behind known failures
that has been introduced with human assistance           may be subject to biased interpretation (Kolar &
outside of its natural range and has successfully        Lodge 2001, García-Berthou et al. 2005, Hayes &
attained a self-sustaining population in the wild.       Barry 2008).
Establishment is one stage in the invasion process,
a sequence commonly described as having four             Although investigations at the global or continental
stages: transport, introduction, establishment,          scale are valuable, such broad studies may be
and spread (Blackburn et al. 2011). In general,          limited in their extrapolative or predictive ability
successful establishment suggests the population         (Gido & Brown 1999, Ruesink 2005). Consequently,
persists, is relatively stable or increasing, and that   studies focused at regional scales can be more
the possibility of its eradication by natural causes     informative in assessing establishment success and
or by humans is unlikely. Establishment failure          invasion risk (Moyle & Marchetti 2006). García-
occurs when a species is unable to overcome              Berthou (2007) reviewed 12 different quantitative
barriers in the transition from introduction to          studies on the characteristics of invasive fishes.
establishment.                                           Among other findings, he concluded that there
                                                         was a lack of comparative studies at multiple
Non-native species that successfully colonize            scales while also noting the absence or shortage
novel areas are often considered to have certain         of research in the tropics and other areas outside
advantageous characteristics, many not mutually          northern temperate regions.
exclusive. In particular, successful invaders are
frequently described as those that are: widespread       The aim of the current study was to identify
and abundant in their native range, habitat or           and assess factors contributing to establishment
Evaluating establishment success of non-native fishes introduced to inland aquatic habitats of tropical Pacific islands
J. Vertebr. Biol. 2021, 70(4): 21064                                   3                   Inland fish introductions on tropical Pacific islands

Fig. 1. Map of the Pacific Ocean showing the locations of the Hawaiian Islands and Guam. The number of established non-native fish
species is included (in parentheses is the number of established species as a percent of all known non-native fishes introduced to inland
fresh waters of each of the two study sites). Many of the larger small tropical islands are shown in red, but thousands of other Pacific
islands are too small to display.

success of non-native freshwater fish species                         The combination of Hawaii and Guam data also
introduced to Hawaii and Guam. Both are oceanic                       permitted us to more adequately evaluate a wider
islands or island groups located in the tropics.                      range of species traits and other factors potentially
Applying a frequentist approach, our analysis                         associated with successful establishment.
focus was on the initial stage of the invasion
process with establishment as the outcome. To                         Study Area
aid our investigation, we also examined data on
fish introductions and establishment for other                        Guam and the Hawaiian Islands both represent
tropical islands in the Pacific and at a global                       small areas of extreme geographic isolation, each
scale. Our research found records documenting                         situated in remote regions of the Pacific Ocean far
as many as 46 non-native, freshwater fish species                     from continental land masses (Fig. 1). The island
to be established in inland and estuarine waters                      of Guam (13.4443° N, 144.7937° E) is part of the
of Hawaii; in contrast, only about 13 species                         Mariana Islands within Micronesia of the western
are known to be established on Guam (Brock                            Pacific Ocean. Guam is the western-most territory
1960, Maciolek 1984, Eldredge 1994, Fuller et al.                     of the United States with a land mass of 547 km2
1999, Yamamoto & Tagawa 2000, Mundy 2005).                            and the largest island in Micronesia. Non-native
Inclusion of Hawaii, rather than Guam alone (our                      freshwater fishes are present in a variety of streams,
original intent), provided necessary and sufficient                   reservoirs, ditches, and ponds on the island. The
data for improving the power of our modelling                         Hawaiian Islands (19.8968° N, 155.5828° W) are an
approach and ultimately allowed development of a                      archipelago of eight major islands and multiple
quantitative model useful for analytical purposes.                    small islets and atolls remotely situated near the
Evaluating establishment success of non-native fishes introduced to inland aquatic habitats of tropical Pacific islands
J. Vertebr. Biol. 2021, 70(4): 21064                                   4                  Inland fish introductions on tropical Pacific islands

Table 1. Model variables pertaining to establishment success of non-native fishes introduced to inland waters of the Hawaiian Islands
and Guam.

    Parameter                    Description                             States
    Dependent                    Status on Hawaii and/or Guam            0 – introduced but not established; 1 – established
    variable
    Independent                  Taxonomic affinity
    variable                        1. Family                            34 different fish families
    Independent                  Distributional attributes and
    variable                     history of introductions
                                       2. History (prior invasion        0-43
                                       success; number of tropical
                                       islands/island groups where
                                       taxon was introduced and
                                       became established)
    Independent                  Propagule pressure
    variable                        3. Pathway                           1 – raised in aquaculture facilities/ponds;
                                                                         2 – in aquarium trade; 3 – stocked as ornamental;
                                                                         4 – biological control; 5 – stocked as food fish;
                                                                         6 – stocked as recreational game fish for sport,
                                                                         commercial, or subsistence fishing; 7 – stocked as
                                                                         prey base; 8 – used as bait; 9 – present in live food
                                                                         markets; 10 – ballast water; 11 – contaminant with
                                                                         stocking of another species; 12 – ceremonial release
    Independent                  Life-history attributes
    variable                         4. Body size (maximum adult 1 – small (< 10 cm); 2 – medium (10-29 cm);
                                     length)                     3 – large (30-100 cm); 4 – very large (> 100 cm)
                                     5. Adult trophic category   1 – detritivore/algivore; 2 – herbivore; 3 –
                                                                 planktivore; 4 – omnivore; 5 – invertivore;
                                                                 6 – invertivore/piscivore; 7 – piscivore/top predator
                                       6. Reproductive guild (level      1 – non-guarders, high fecundity, small egg;
                                       of parental care, fecundity,      2 – guarder, medium fecundity and egg size;
                                       and egg size)                     3 – bearer, low fecundity and large egg sizea

                                       7. Spawning habitat               1 – lotic; 2 – lentic; 3 – coastal/estuarineb;
                                                                         4 – lotic and lentic; 5 – lotic, lentic and coastal/
                                                                         estuarine
    Independent                  Physiological tolerance
    variable                        8. Climate profile (climate          1 – temperate; 2 – subtropical; 3 – tropical; 4 –
                                    regions included in native           temperate/subtropical; 5 – subtropical/tropical
                                    geographic distribution)

                                       9. Salinity tolerance             1 – intolerant to salinity < 1 ppm (stenohaline);
                                                                         2 – moderately tolerant > 1 and < 10 ppm; 3 – very
                                                                         tolerant > 10 ppm (euryhaline)
                                       10. Hypoxia tolerance             1 – intolerant of low dissolved oxygen;
                                                                         2 – moderately tolerant; 3 – tolerant, air breathing
                                       11. Environmental tolerance       3-12 (summation of scores for variables 8-10)
a
    Includes live-bearing species (e.g. Poecilia), species that carry eggs externally (e.g. Oryzias, Ancistrus), and species that brood eggs
    internally in their orobranchial cavity (e.g. Oreochromis, Melanochromis).
b
    Includes species that spawn offshore or on reefs (e.g. Anguilla marmorata).
J. Vertebr. Biol. 2021, 70(4): 21064                    5               Inland fish introductions on tropical Pacific islands

centre of the Pacific Ocean (Juvik & Juvik 1998; Fig.   cavifrons) and highly euryhaline species that
1). Its nearest major landmass is North America         inhabit fresh, brackish, and marine waters (e.g.
about 5,000 km to the east. The islands are a state     Gulf killifish Fundulus grandis and Mozambique
of the United States and have a combined land area      tilapia Oreochromis mossambicus).
of about 16,640 km2. Non-native freshwater fishes
have been introduced to at least seven of the major     Positive identification of non-native fish species
islands of Hawaii, the greatest number occurring        is a challenge, especially given the taxonomic
on the island of Oahu (Maciolek 1984, USGS 2021).       instability of certain groups and growing
Non-native fishes are found in most inland aquatic      recognition of cryptic forms. During our
habitats of Hawaii, including various streams,          assignment of fish names, we relied heavily on
reservoirs, natural and artificial wetlands and         recent taxonomic revisions and occasionally
lagoons, Hawaiian fishponds (ancient aquaculture        consulted with taxonomic experts. In the current
structures) and farm ponds, ditches, and natural        study, 14 species identifications were questionable,
anchialine pools, as well as various estuarine sites    representing taxa that were originally reported
(Yamamoto & Tagawa 2000, Mundy 2005).                   as genus only, belong to unresolved taxonomic
                                                        complexes, or are commonly misidentified
Material and Methods                                    cryptic species. For these taxa, variables used in
                                                        the analysis were scored based on species most
Non-native fishes included in estimation of             likely to have been introduced as evidenced from
establishment success                                   other introductions (Maciolek 1984, Fuller et al.
Information on Pacific Island fish introductions        1999, Eldredge 2000, Mundy 2005, USGS 2021),
and establishment status used in this study were        or putative related species presumed to share
gathered from many sources. Included were data          similar life-history, ecological, and physiological
assembled from an in-depth review of published          attributes. The species and status of all taxa in our
and unpublished literature, consultation with           data set are summarized in Table S1.
numerous experts and local authorities, search of
museum records and specimens, and examination           Variables
of records in the U.S. Geological Survey’s              Ten independent variables were selected a priori to
Nonindigenous Aquatic Species (NAS) database            evaluate establishment success: 1) family; 2) number
(USGS 2021). A few of the more useful sources           of tropical islands or tropical island groups where
of information on non-native fishes in the Pacific      a species was introduced and became established,
region are works that contain review and summary        excluding Hawaii and Guam (Table S2); 3) pathway
data (e.g., Maciolek 1984, Devick 1991, Eldredge        of introduction or human use; 4) maximum body
1994, 2000, Fuller et al. 1999, Yamamoto & Tagawa       length; 5) adult trophic category; 6) reproductive
2000, Mundy 2005). In addition, we relied heavily       guild; 7) spawning habitat; 8) climate in a species’
on our own original field observations including        native range; 9) salinity tolerance, and; 10) hypoxia
fish specimens collected by us during surveys           tolerance (Table 1). An additional independent
conducted in Hawaii, Guam, Pohnpei, Saipan, the         variable (11), designated environmental tolerance,
Republic of Palau, and the Galapagos.                   was established a posteriori and represented a
                                                        composite of variables 8-10. These variables were
Guam was treated as analogous to a Hawaiian             scored based on published literature, technical data
island in the context of evaluating establishment       summarized in FishBase (Froese & Pauly 2021),
success of non-native fishes among islands of the       closely related or similar species in cases where
broad study area of the tropical Pacific. Thus,         data were unavailable, and personal knowledge of
data for both regions (81 taxa, both introduced         the authors. Detailed descriptions and rationale for
and established) were combined for analysis,            selection of variables are provided in Appendix S1.
which was required to obtain model convergence.
An established species was defined as any taxon         Model set
known to have one or more self-sustaining wild          The data matrix used in our model of establishment
populations in Hawaii, Guam, or both. Species           success is provided in Table S3. Establishment
include those that are restricted to fresh water        success was a binary response variable scored as 1
in their native distribution (most species in our       for non-native fish species with one or more self-
data matrix), species that occur in both fresh and      sustaining, reproducing populations in the wild in
brackish waters (e.g. mangrove goby Mugilogobius        Hawaii or Guam, and 0 for non-native fish species
J. Vertebr. Biol. 2021, 70(4): 21064                                  6                  Inland fish introductions on tropical Pacific islands

Table 2. Set of competing models of establishment success for non-native fishes introduced into inland waters of Hawaii and Guam and
the hypotheses that each model represents. The first 21 models were defined a priori. Models 22-28 were a posteriori variation of the a
priori model with the most support. Model 29 was an a posteriori version of the best model among models 1-28 in which the number of
hypoxia tolerance scores were reduced from 3 to 2.

 Model
             Model name                                         Hypothesis                                 Reference
 No.
 1           Family + history + pathways + length +  Our most general model                                Present study
             diet
             + reproductive guild + spawning habitat
             + environmental tolerance
 2           Family + history + pathways + length +             Best model of Marchetti et al.             Marchetti et al. 2004a
             diet                                               2004a
             + reproductive guild + environmental
             tolerance
 3           Family + pathways + length + diet                  Invasive species characteristics Kolar & Lodge 2001
             + reproductive guild + environmental
             tolerance
 4           Family + history + pathways + length               Expert opinion model                       Marchetti et al. 2004a
             + reproductive guild + environmental
             tolerance
 5           Family + history + pathways + diet                 Human interest and habitat                 Present study
             + spawning habitat + environmental                 generalist
             tolerance
 6           Family + history + pathways +                      Best model in a different study Marchetti et al. 2004b
             reproductive guild + environmental
             tolerance
 7           Family + length + diet + reproductive              Life history                               Present study
             guild
             + spawning habitat
 8           Family + pathways + diet +                         Invasive species characteristics Moyle & Light 1996a, b
             environmental tolerance
 9           Family + diet + reproductive guild                 Best model 1 of Kolar & Lodge Kolar & Lodge 2002
             + environmental tolerance                          2002
 10          Family + history + reproductive guild              Best model 2 of Kolar & Lodge Kolar & Lodge 2002
             + environmental tolerance                          2002
 11          Family + pathways + reproductive guild             Propagule pressure,           Williamson & Fitter
             + spawning habitat + environmental                 population growth and habitat 1996a, b
             tolerance
 12          Family + diet + spawning habitat                   Habitat generalist                         Present study
             + environmental tolerance
 13          Family + diet + environmental tolerance            Ecological characteristics or              Marchetti et al. 2004a
                                                                novel environments
 14          Family + length + reproductive guild               Life history or high population Marchetti et al. 2004a
                                                                growth
 15          Family + history + pathways                        Human interest                             Marchetti et al. 2004a
 16          Family + pathways + diet                           Best model in a different study Ruesink 2005
 17          Family + history + spawning habitat                Prior establishment success                Present study
                                                                and spawning habitat
 18          Family + history + length                          Best model in a different study Ribeiro et al. 2008
 19          Family + history                                   Taxonomic affinity + Prior                 Bomford et al. 2010
                                                                establishment success
J. Vertebr. Biol. 2021, 70(4): 21064                       7                Inland fish introductions on tropical Pacific islands

Table 2. continued

 Model
              Model name                               Hypothesis                              Reference
 No.
 20           Family + pathways                        Propagule pressure                      Blackburn et al. 2009
 21           Family                                   Taxonomic effect only                   Present study
 22           Family + history + climate profile       Best a priori model + climate           Bomford et al. 2010
                                                       profile
 23           Family + history + hypoxia tolerance     Best a priori model + hypoxia           Present study
                                                       tolerance
 24           Family + history + salinity tolerance    Best a priori model + salinity          Present study
                                                       tolerance
 25           Family + history + climate profile +     Best a priori model + climate           Present study
              hypoxia tolerance                        profile + hypoxia tolerance
 26           Family + history + climate profile +     Best a priori model + climate           Present study
              salinity tolerance                       profile + salinity tolerance
 27           Family + history + hypoxia tolerance +   Best a priori model + hypoxia           Present study
              salinity tolerance                       tolerance + salinity tolerance
 28           Family + history + climate profile +     Best a priori model + climate           Present study
              hypoxia tolerance + salinity tolerance   profile + hypoxia tolerance +
                                                       salinity tolerance
 29           Family + history + hypoxia tolerance     Best a priori model + hypoxia           Present study
                                                       with number of hypoxia
                                                       scores reduced from 3 to 2
                                                       (hypoxia scores of 1 and 2
                                                       lumped vs. 3)

introduced but not known to be established in those       could be reproduced exactly using our variable
two areas. Introduction records and establishment         set. In such cases we approximated published
success were based on published or online sources         models as closely as possible with the variables in
(Maciolek 1984, Eldredge 1994, 2000, Fuller et al.        our dataset. A few additional competing models
1999, Mundy 2005, USGS 2021). Logistic regression         were added based on consideration of our unique
was used for our analysis because of the binary           dataset.
nature of our response variable (Agresti 2013, 2018).
This approach is common in studies of establishment       Each of the 21 a priori models in our model set
success of non-native species (Marchetti et al. 2004a,    were ranked using Akaike’s Information Criterion
b, Ruesink 2005, Ribeiro et al. 2008, Bomford et al.      (AICc) adjusted for small sample size (Burnham &
2009, 2010, Fujisaki et al. 2009).                        Anderson 2002, Symonds & Moussalli 2011). AICc
                                                          balances model precision and bias; the best model
We created a set of 21 competing a priori logistic        has the lowest AICc. We calculated Akaike weights
regression models each representing a different           (wi) to evaluate support for each model and for
hypothesis regarding establishment success                individual variables. These Akaike weights sum to
(Table 2). This model set included the most               1.0 and indicate relative support for a given model.
general model that contained as many of the               Climate, salinity tolerance, and hypoxia tolerance
ten independent variables as possible while still         were added to the best model in an exploratory
achieving convergence. The most general model             post-hoc fashion by creating seven additional
included the composite environmental tolerance            models to represent every possible combination
variable. The remaining 20 models in the set each         of these three variables. An additional a posteriori
included a unique subset of the independent               model was added that reduced the number of
variables in the most general model. Most of the          hypoxia tolerance scores from three to two (hypoxia
competing models were formulated by reviewing             intolerant and moderately tolerant combined vs.
the literature. Not all previously published models       air breathing).
J. Vertebr. Biol. 2021, 70(4): 21064                                  8                   Inland fish introductions on tropical Pacific islands

Model approach                                                      A two-sided P-value was used to estimate whether
To specify differences in establishment among                       the test statistic differed significantly from values
species, the probability of establishment on the                    obtained via simulation. Results of the model
logit scale was formulated as a linear combination                  fitting procedure are summarized in Appendix S2
of random and fixed covariates. Model parameters                    and Fig. S1 and S2.
were estimated using the “lmer” function for
linear mixed models (Gelman & Hill 2006, Gałecki                    Results
& Burzykowski 2013, West et al. 2015) in R (R
Core Team 2020). Each of the 21 a priori competing                  Of the 81 non-native fish taxa included in this
models included a random family effect. When no                     analysis, 50 (62%) became established in Hawaii,
random effect is supported by the data the “lmer”                   Guam, or to both places, and 31 (38%) failed to
function estimates a constant intercept with a                      become established. Fifteen taxa were introduced
variance of zero, essentially removing the random                   to both Hawaii and Guam. Number of known
family effect from the model. Twenty-one families                   established species for Hawaii is 46 and for Guam
were represented by a single species; inclusion                     is 13. Ten taxa became established in both the
of families with only one species in our model is                   Hawaiian Islands and Guam. The non-native fishes
not a problem when using a random family effect                     in our analysis represent 33 families and 64 genera
(Gelman & Hill 2006, Gelman et al. 2020).                           (Table S1). The family Cichlidae was represented
                                                                    by the most species (18), followed by Poeciliidae
After identifying the best model, we assessed fit                   (8), Cyprinidae (7), Centrarchidae (4), Loricariidae
of that model by estimating its error rate. Error                   (4), and Salmonidae (4). All other families were
rate is the proportion of observations in the data                  represented by one to three species introduced.
set for which the fitted value > 0.5 and species                    Numbers of non-native fish taxa introduced and
status in the study area = 0 or fitted value < 0.5 and              established in inland waters of the Hawaiian
status = 1. We compared the error rate of the best                  Islands, Guam, other tropical Pacific islands or
model to the error rate of the null model. The null                 island groups, and all tropical islands globally are
model is simply the proportion of observations                      summarized in Table 3.
in the data set for which species status = 1. Error
rate of the null model is 1 minus the proportion of                 Just over one-half (41 species) of the non-native
observations in the data set with a species status                  fishes introduced to Hawaii or Guam are also
= 1. Ideally the error rate of the best model is low                established globally on other tropical islands/
and substantially lower than the error rate of the                  island groups. Excluding Hawaii and Guam,
null model.                                                         number of islands/groups on which a species
                                                                    became established ranged from 0 to 43 (mean =
We also assessed fit of the most general model by                   3.2). Of the 81 species in our data set, 46 (58%) were
creating 1,000 simulated data sets based on that                    introduced to other tropical Pacific islands where
model using the “sim” function in the R package                     they did not become established, and 39 (49%) were
“arm” (Gelman & Hill 2006, Gelman et al. 2020).                     introduced to other tropical islands globally but
We used proportion of observations in the data set                  did not become established. Eight species became
with a species status = 0 as our test statistic and                 established on one or more tropical Pacific islands/
compared the value of that test statistic to the same               groups, and 12 species on tropical islands/groups
proportion for each of the 1,000 simulated data sets.               globally, but not in the Hawaiian Islands or Guam.

Table 3. Summary of the numbers of introduced and established non-native fish taxa in inland waters of the Hawaiian Islands and Guam.
Values in last two columns represent number of taxa in common with those introduced (established or not) on Hawaiian Islands, Guam,
or both. Additional non-native fish species never introduced to Guam or Hawaii but established on other tropical islands are excluded.

                                                  Number of taxa             Number of taxa
                                   Number                                                                     Number of taxa
   Island or island                                 established              established on
                                    of taxa                                                                 established on other
        group                                   (percentage of total          other tropical
                                  introduced                                                              tropical islands globally
                                                    introduced)              Pacific islands
         Hawaii                         70            46 (66%)                        31                                   37
          Guam                          25            13 (52%)                        15                                   18
 Hawaii plus Guam                       81            50 (62%)                        35                                   41
J. Vertebr. Biol. 2021, 70(4): 21064                                9                  Inland fish introductions on tropical Pacific islands

The following numbers and percentages pertain                      All 81 non-native fishes introduced to Hawaii,
to species introduced to tropical islands/island                   Guam, or both places were scored for each of the 13
groups globally. Thirteen species (16%) became                     pathway categories (Fig. S3). Sixty-three of the 81
established on only one other tropical island/group,               fish taxa were recorded as being in the aquaculture
and 24 (30%) became established on from 2-12                       industry pathway and of these 63% became
tropical islands/groups. A few species are widely                  established. Numbers of species and percentage of
established among different tropical islands/                      these established for the other pathways were: 57
groups. The most broadly distributed include the                   (68%) present in the aquarium trade; seven (86%)
Mozambique tilapia, recorded as established on                     stocked as ornamentals; 22 (64%) used as agents for
at least 43 other tropical islands/groups, and three               biological control; 20 (55%) stocked or transported
different members of the family Poeciliidae: the                   as fish for human consumption; 20 (60%) stocked
guppy Poecilia reticulata established on 30 other                  as sport or game fish; six (100%) stocked as
tropical islands/groups; the mosquitofish Gambusia                 forage for other fish; 11 (73%) used as bait; 23
spp. – Gambusia affinis and Gambusia holbrooki                     (61%) present in live food markets; two (100%)
identifications are often commingled (Courtenay                    transported in ballast water; five (20%) occurring
& Meffe 1989, Fuller et al. 1999) – established on                 as unintentional contaminants with, or in place
23 other tropical islands/groups; and the green                    of, other species, and; one (100%) known to be
swordtail Xiphophorus hellerii established on 15                   associated with ceremonial release. No species was
other tropical islands/groups. All four of these taxa              scored as having been introduced for conservation
are also established in Hawaii and Guam. Twenty-                   purposes. Although rock flagtail Kuhlia rupestris
two of 23 species introduced into Hawaii or Guam                   has been stocked in Australia to restore populations
and established on at least three other tropical                   (Hutchison et al. 2002), in this study it was not
islands/groups became established in Hawaii or                     scored as such because its introduction to Hawaii,
Guam. The giant goramy Osphronemus goramy                          where it failed to establish, was likely for sport
was established on seven other tropical islands/                   fishing (Brock 1960). The total number of pathways
groups, but did not become established after being                 any single species scored ranged from one to eight
introduced into Hawaii (Brock 1960, Mundy 2005).                   (mean = 2.98). The common carp Cyprinus carpio

Table 4. Best models of establishment success of non-native fishes introduced to inland waters of Hawaii and Guam. Only models with
Akaike weight ≥ 0.01 are included. Asterisks designate a priori models. Model 29 was excluded from comparison since that post-hoc
model was two steps removed from being a priori.

                                                             Number of
  Model                                                                            Log-                                           Akaike
        Model                                               parameters in                           AICc           Δ AICc
 number                                                                         likelihood                                        weight
                                                               model
     23        Family + history + hypoxia tolerance                 5            –39.2382        89.28725               0           0.587
     27        Family + history + hypoxia tolerance
                                                                    7            –38.7221        92.9998          3.712554          0.092
               + salinity tolerance
    19*        Family + history                                     3            –43.4619        93.23956         3.952312          0.081
     25        Family + history + climate profile +
                                                                    7            –38.9803        93.51617         4.228921          0.071
               hypoxia tolerance
    10*        Family + history + reproductive guild
                                                                    6            –40.7468        94.64419         5.356942          0.040
               + environmental tolerance
    15*        Family + history + pathways                          4            –43.1532        94.83976         5.552514          0.037
    18*        Family + history + length                            6            –40.8552        94.86112         5.573872          0.036
     24        Family + history + salinity                          5            –42.7962        96.40313         7.115879          0.017
     6*        Family + history + pathways +
               reproductive guild + environmental                   7            –40.6548        96.86508         7.577831          0.013
               tolerance
     22        Family + history + climate profile                   5            –43.2218        97.25438         7.967138          0.011
     28        Family + history + climate profile +
                                                                    9            –38.6093        97.79007         8.502822          0.008
               hypoxia tolerance + salinity tolerance
J. Vertebr. Biol. 2021, 70(4): 21064                                   10                 Inland fish introductions on tropical Pacific islands

and Mozambique tilapia were associated with the                       Guam. According to our best model air-breathing
greatest number of different pathways, each with                      fishes were most likely to become established
the maximum of eight categories.                                      (β,ˆ hypoxia 3 = 2.36, SÊ = 1.71). Fourteen air-breathing
                                                                      species were included in our study; of these, 11
Models of establishment success of non-native                         (79%) are established in Hawaii or on Guam, but
fish species introduced to Hawaii and Guam                            only the snakehead Channa spp. is common to both
were compared using an information-theoretic                          (Table S1).
approach. The best a priori model (19), which was
the third best model overall, included an effect of                   Point estimates for the random family effect in our
taxonomic affinity (family) and prior establishment                   best model ranged from –3.48 to 3.76 (Table S4).
success (history) on tropical islands/island groups                   Of the nine families represented by two species or
excluding Hawaii and Guam (Table 4). The best                         more, Cichlidae (3.76), Centrarchidae (2.54), and
overall model of establishment probability, which                     Poeciliidae (2.11) had the highest point estimates,
was a post-hoc model (23), included a random                          representing relatively high establishment
family effect with a non-zero standard deviation                      probability among the families represented in
(SD = 2.61), an effect of prior establishment                         our data set. Osphronemidae (–3.48), Salmonidae
success on tropical islands/island groups, and an                     (–1.76), and Anguillidae (–0.88) had the lowest point
effect of hypoxia tolerance (Table 4). This model                     estimates, indicating relatively low establishment
with the lowest AICc and highest Akaike weight                        probability. These establishment probabilities at
had substantial support (Akaike weight = 0.59,                        the family level are relative to each other since they
evidence ratio = w25/w27 = 6.4). No other model                       are drawn from a common distribution.
had an Akaike weight > 0.1 and the second, third
and fourth best models were only minor variations                     We plotted probability of establishment based
of our best model. Effect of prior establishment                      on our best model using the mean of fixed plus
success in our best model had a positive slope,                       random family intercept (Kinney & Dunson
as expected (β,ˆhistory = 0.45, SÊ = 0.19), suggesting                2007) that incorporated the history variable as the
establishment probability in Hawaii and Guam was                      x-axis (Fig. 2). The mean estimated establishment
higher for species established on a large number                      probability was 0.37 among all 13 species with
of islands/island groups excluding Hawaii and                         a hypoxia score of one and not established on other

             Fig. 2. Probability of establishment of non-native fishes introduced to inland waters of Hawaii and Guam as a
             function of number of tropical islands/island groups on which a species is established, excluding Hawaii and
             Guam, and hypoxia tolerance. Logit-scale plots based on mean of fixed plus random family intercept. Hypoxia
             tolerance score: 1 – intolerant of low dissolved oxygen, 2 – moderately tolerant, 3 – are air-breathers. Closed
             circles represent number of tropical islands or island groups on which a species is known to be established;
             many circles represent more than one species.
J. Vertebr. Biol. 2021, 70(4): 21064                                 11                 Inland fish introductions on tropical Pacific islands

islands/island groups, using estimated family                       established on six or more islands/island groups
intercepts (estimated probability was 0.44 using                    using a fixed intercept (mean estimated probability
a fixed intercept). Mean estimated probability of                   > 0.92 using a random intercept). For eight species
establishment was > 0.30 for species with a hypoxia                 established on ≥ 10 islands/island groups, all with
score of one if established on two or more islands/                 a hypoxia score of two, mean estimated probability
island groups using random family intercepts                        of establishment was 0.86 using the fixed intercept
(estimated probability > 0.66 using fixed intercept),               (0.99 using random family intercepts). Species with
but only three species were in this category in the                 a hypoxia tolerance score of three (air breathers)
data set (the brown trout Salmo trutta, the channel                 had an 0.89 estimated probability of becoming
catfish Ictalurus punctatus, and the rainbow trout                  established if not established on any other island/
Oncorhynchus mykiss). Species with a hypoxia                        island group, also using a fixed intercept; the
tolerance score of two had only a 0.06 probability of               mean estimated probability for the seven species
becoming established if they were not established                   of this group was marginally lower, 0.84, when
on any other island/island group using a fixed                      using random family intercepts. For air-breathing
intercept. However, when using random family                        species established on ≥ 3 islands/island groups,
effects, the mean estimated probability was 0.39 for                the estimated probability was 0.97 using a random
the 19 species of this group. The probability rose to               family intercept and reached ≥ 0.99 for those on ≥ 6
> 0.49 for species with a hypoxia score of two and                  islands/island groups.

Fig. 3. Probability of establishment of non-native fishes of selected families introduced to inland waters of Hawaii and Guam as a
function of number of tropical islands/island groups on which a species is established, excluding Hawaii and Guam, hypoxia tolerance,
and family effect used as intercept. N – number of species for each family/hypoxia score combination.
J. Vertebr. Biol. 2021, 70(4): 21064                    12                                                                                                                                                                                                                                            Inland fish introductions on tropical Pacific islands

                                                          Table 5. Estimated establishment probability of selected fish taxa not known to be currently established in Hawaii or on Guam using parameters of best-fit model (#23). Shading indicates taxa on Guam
We also plotted estimated probability of

                                                                                                                                                                                                                                                                   deviation
                                                                                                                                                                                                                                                                   Standard
establishment for selected families with emphasis

                                                                                                                                                                                                                                                                                                                                                                                                                                                                    0.36
                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      0.11
                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               0.11
                                                                                                                                                                                                                                                                                                                                                                                                               0.37
                                                                                                                                                                                                                                                                                                                                                                                                                                 0.36
                                                                                                                                                                                                                                                                                                                                                     0.35
                                                                                                                                                                                                                                                                                                                                                                               0.37
                                                                                                                                                                                                                                                                                            0.11
                                                                                                                                                                                                                                                                                                                            0.17
of those represented by more than three species
in our data set (Fig. 3). Members of the family
Cyprinidae with a hypoxia score of one had

                                                                                                                                                                                                                                                                   Establishment
                                                                                                                                                                                                                                                                    probability
an establishment probability of 0.55 when not

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               0.07
                                                                                                                                                                                                                                                                                                                                                                                                                                                                    0.25
                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      0.07
                                                                                                                                                                                                                                                                                                                                                                                                               0.75
                                                                                                                                                                                                                                                                                                                                                                                                                                 0.74
                                                                                                                                                                                                                                                                                                                                                     0.77
                                                                                                                                                                                                                                                                                                                                                                               0.75
                                                                                                                                                                                                                                                                                            0.95
                                                                                                                                                                                                                                                                                                                            0.92
established on other islands/island groups. For
cyprinids with a hypoxia score of two, probability
was < 0.13 when not established on other islands/
island groups but was 0.96 when established on 12

                                                                                                                                                                                                                                                                   tolerance score
or more islands/island groups, due to widespread

                                                                                                                                                                                                                                                                      Hypoxia
introductions of the goldfish Carassius auratus,
and the common carp. Species of the family

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      2
                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               2
                                                                                                                                                                                                                                                                                                                                                                                                               3
                                                                                                                                                                                                                                                                                                                                                                                                                                 3

                                                                                                                                                                                                                                                                                                                                                                                                                                                                    2
                                                                                                                                                                                                                                                                                                                                                     3
                                                                                                                                                                                                                                                                                                                                                                               3
                                                                                                                                                                                                                                                                                            3
                                                                                                                                                                                                                                                                                                                            3
Salmonidae, all with hypoxia scores of one, had
a probability of 0.15 when not established on
other islands/island groups, with the probability

                                                                                                                                                                                                                                                                   Tropical island groups
                                                                                                                                                                                                                                                                    on which established
increasing to 0.30 and 0.63 when established on
two or five island/island groups, respectively.
The family Poeciliidae, all species with hypoxia
score of two, had a probability of 0.39 when not

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      0
                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               0
                                                                                                                                                                                                                                                                                                                                                                                                                                 2

                                                                                                                                                                                                                                                                                                                                                                                                                                                                    1
                                                                                                                                                                                                                                                                                                                                                                               0
                                                                                                                                                                                                                                                                                                                                                                                                               0
                                                                                                                                                                                                                                                                                            0
                                                                                                                                                                                                                                                                                                                            0
                                                                                                                                                                                                                                                                                                                                                     1
established on other islands/island groups, but
the probability increased to 0.86 and 0.99 when
established on five or ten island/island groups. Air-
breathing members of the Loricariidae had a 0.93
probability of establishment when not established

                                                                                                                                                                                                                                                                                                                                                                                                                                 Australasian/African bonytongues
on other island/island groups, and this probability
increased to 0.99 when established on three island/
                                                                                                                                                                                                                                                                                            Stinging or air-sac catfishes

island groups; no loricariids with a hypoxia score of
two were reported to occur on other island/island
groups. The Centrarchidae (hypoxia scores of one

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               Fathead minnow
and two) had high establishment probabilities,
                                                                                                                                                                                                                                                                        Common name

                                                                                                                                                                                                                                                                                                                                                     Featherbacks
                                                                                                                                                                                                                                                                                                                            Brown hoplo

ranging from 0.71-0.94 if established on two to

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      Black carp
                                                                                                                                                                                                                                                                                                                                                                                                                                                                    Spiny eels
six other islands/island groups and reaching 0.99
                                                          Division of Aquatic and Wildlife Resources white list permitted for importation.

                                                                                                                                                                                                                                                                                                                                                                                                               Bichirs

if established on ten island/island groups. The
                                                                                                                                                                                                                                                                                                                                                                               Gars

Cichlidae, all with hypoxia score of two, also had
high estimated probabilities of establishment
ranging from 0.77 if not established on any other
                                                                                                                                                                                                                                                                                                                                                                                                                                                                    Mastacembelus/Macrognathus spp.

islands/island groups, increasing to 0.84, 0.93 and
0.99 if established on no, one, or five other islands/
                                                                                                                                                                                                                                                                                                                                                                               Lepisosteus/Atractosteus spp.

island groups, respectively.
                                                                                                                                                                                                                                                                                                                                                                                                                                 Scleropages/Heterotis spp.

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      Mylopharyngodon piceus
                                                                                                                                                                                                                                                                                                                                                     Notopterus/Chitala spp.
                                                                                                                                                                                                                                                                                                                            Hoplosternum littorale

The air-breathing family Channidae had an
                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               Pimephales promelas
                                                                                                                                                                                                                                                                                            Heteropneustes spp.

estimated establishment probability of 0.90 when
                                                                                                                                                                                                                                                                        Scientific name

                                                                                                                                                                                                                                                                                                                                                                                                               Polypterus spp.

not established elsewhere (as previously noted, the
two snakehead species reported for Hawaii and
Guam were treated as a single taxon in our model).
Species of the Channa complex were established on
seven other tropical islands/island groups and the
fitted value of establishment probability for this
                                                                                                                                                                                                                                                                                                                                                                                                                                                                    Mastacembelidae

taxon was 0.99 (Table S1).
                                                                                                                                                                                                                                                                                                                                                                                                                                 Osteoglossidae/
                                                                                                                                                                                                                                                                                                                            Callichthyidae

                                                                                                                                                                                                                                                                                                                                                                               Lepisosteidae
                                                                                                                                                                                                                                                                                                                                                     Notopteridae

                                                                                                                                                                                                                                                                                                                                                                                                                                 Arapaimidae
                                                                                                                                                                                                                                                                                                                                                                                                               Polypteridae

                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                      Cyprinidae
                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                                               Cyprinidae

Using our best-fit model we scored parameters
                                                                                                                                                                                                                                                                                            Clariidae
                                                                                                                                                                                                                                                                        Family

for a set of nine taxa not known to be currently
established in Hawaii or Guam to assess possibility
J. Vertebr. Biol. 2021, 70(4): 21064                  13               Inland fish introductions on tropical Pacific islands

of future establishment (Table 5). These taxa were    that can survive and reproduce in fresh, brackish,
selected based on their known or presumed             and marine environments and may occasionally
likelihood of presence in pathways of concern,        make forays into reef habitats (Lobel 1980, Nelson
primarily popularity or novelty in the aquarium       & Eldredge 1991). Nearly all the fish species
trade and species used for aquaculture, food, bait,   introduced to Hawaii and Guam belong to families
or as biocontrol agents. We also sought to include    not represented by native species on these islands.
some species that are on a white list maintained
by the Guam Division of Aquatic and Wildlife          Not unexpectedly, the family-level composition
Resources, which are species that are permitted       of many fish species introduced and successful
for importation (Miller 2014). The establishment      at becoming established in Hawaii and Guam
probability for six of the taxa was greater than      includes groups of substantial human value and
70% with two greater than 90%. The remaining          interest, especially commercially important food
three taxa had an establishment probability of        species raised in aquaculture, sport fishes, species
25% or less. There was considerable imprecision       in the ornamental trade, and species introduced for
in some of these estimations given the relatively     biocontrol (Maciolek 1984, Eldredge 1994, 2000).
high standard deviations of predicted probability     This taxonomic and economic bias is common in
of establishment. This was primarily due to greater   studies of introduced fishes in other parts of the
relative importance of hypoxia tolerance scores       world (Rahel 2002, 2007, Alcaraz et al. 2005, Jeschke
and low numbers of tropical islands/island groups     & Strayer 2005, 2006, García-Berthou 2007, Strayer
on which the species are established (none for six    2010), and such economic-linked introduction
taxa, one for two taxa, and two for one taxon).       pathways are common for freshwater fishes in
Nevertheless, this procedure illustrates how the      general (Fuller 2003, García-Berthou et al. 2005,
modelling method can be applied to provide a          Kerr et al. 2005, Gertzen et al. 2008, Jiao et al. 2009,
general estimate of establishment likelihood for      Strecker et al. 2011, Lapointe et al. 2016, García-
taxa of possible concern.                             Díaz et al. 2018, Piria et al. 2018). Numerous species
                                                      that became established in Hawaii, on Guam, and
Discussion                                            many other tropical islands in the Pacific and
                                                      globally were legal importations and many were
In this study the best probability model of           intentionally introduced into the wild, often by
establishment success included: a phylogenetic        natural-resource management agencies or entities
component, indicated by the family to which           associated with commercial interests (Nelson &
a species belongs; a history component,               Eldredge 1991, Fuller et al. 1999, Eldredge 2000,
characterized by the number of occurrences in         Yamamoto & Tagawa 2000). Introductions of fishes
which a species previously became established         and other freshwater species on tropical Pacific
on other tropical islands/island groups globally;     islands fall into major chronological periods
and a categorical score of tolerance to low levels    associated with development and expansion of
of dissolved oxygen (i.e. hypoxia tolerance). Our     human activities in the region (Maciolek 1984,
modelling results were generally consistent with      Eldredge 1992, 1994, 2000). Although introductions
aspects of numerous other studies that examined       of freshwater fishes to Pacific Islands began in the
non-native fish establishment success relative to     late 1800s, most introductions occurred from the
historical factors and trait-based characters.        mid-1900s to the present. It is easiest to identify
                                                      intentional introductions based on many records
Like other oceanic islands with permanent             for which good sources of published information
freshwater habitats, Hawaii and Guam have             exist. However, in many cases information is
depauperate native freshwater fish faunas             lacking, especially in cases where introductions
(Pippard 2012) and most indigenous species            or establishment were unintentional. We surmise
have an amphidromous or catadromous life cycle        that most cases of unintentional establishment of
involving life-history stages requiring both fresh    freshwater fishes on tropical Pacific islands were
and salt water (McDowall 1988, 2004). In contrast,    the result of released aquarium fish, escapement
most fish species introduced to these islands are     from captive holding facilities, and contaminants
strictly freshwater in their native environments.     associated with intentional introductions. Thus,
However, there are notable exceptions, such as        although there are far fewer studies of freshwater
the euryhaline Mozambique tilapia, a species          fish introductions to tropical island regions
J. Vertebr. Biol. 2021, 70(4): 21064                      14               Inland fish introductions on tropical Pacific islands

compared to continental faunas, our results               In terms of pathways, the origins, sources,
corroborate that many of the same factors operate         purposes, approximate dates, and general
globally in contributing to successful establishment      fate of most intentional introductions of non-
of freshwater fishes in novel regions.                    native fishes to inland habitats of Pacific islands,
                                                          including Hawaii and Guam, were reviewed in
Propagule pressure is widely regarded as a primary        many publications from which we assembled our
determinant of establishment success of non-              data matrix (Brock 1960, Maciolek 1984, Devick
native species and is a driver for other stages of the    1991, Eldredge 1994, 2000, Englund 2002). As
invasion process (Kolar & Lodge 2001, Leung et al.        noted, intentional introductions of non-native
2004, Lockwood et al. 2005, 2009, Colautti et al. 2006,   fishes to inland waters of Hawaii and Guam were
Drake & Lodge 2006, Simberloff 2009, Brockerhoff          primarily to stock sport or game species or their
et al. 2014). Herein, we did not have a specific          prey, establish breeding populations of food fish
measure of propagule pressure, given uncertainty          in aquaculture facilities or natural water bodies,
or lack of data for most species about total              and biocontrol. This is reflected by several of the
numbers of individuals introduced, frequency of           families and their respective species listed in Table
introductions, and numbers of sites where released.       S1. For example, many of the species introduced
However, we consider two conditions in our study          for sport or game fishing include cichlids (Cichla
to be indirect indicators of relatively high propagule    ocellaris), centrarchids (Micropterus spp., Lepomis
pressure and successful fish establishment: number        spp.), ictalurids (Ameiurus nebulosus, I. punctatus),
of tropical islands/island groups on which species        and salmonids (Oncorhyncus mykiss, Oncorhyncus
were previously established, and number of                tshawytscha, S. trutta, Salvelinus fontinalis). Examples
different pathways through which various species          of species stocked to serve as forage base for other
are transported. Many of the non-native fish species      fishes, both native and introduced, include clupeids
that successfully colonized Hawaii and Guam are           (Dorosoma petenense), centarchids (Lepomis spp.)
also established on many other tropical island/           and poeciliids (Xiphophorus spp.). Included among
island groups globally. The historical component is       the many species introduced into inland waters
obviously important and, as such, our best logistic       either purposefully or as unintended releases
model is congruent with other studies of freshwater       in association with aquaculture or intention to
fishes that identify prior invasion success as a          establish populations for harvest as food were
reliable indicator of successful establishment in         cyprinids (C. carpio, C. auratus, Hypophthalmichthys
other regions. Prior invasion success generally           nobilis), clariids (Clarias spp.), cichlids (Oreochromis
represents a human-interest component (Jeschke &          spp., Tilapia spp.), and channids (Channa spp.).
Strayer 2005, 2006). In the Laurentian Great Lakes        Species introduced for biocontrol, typically to
of North America, introduced fishes with a prior          manage invasive plants, phytoplankton, other
history of invasiveness outside of their native range     fishes, and mosquito control include cyprinids
were more likely to become established (Kolar &           (Ctenopharyngodon idella), cichlids (Oreochromis
Lodge 2002, Rixon et al. 2005, Snyder et al. 2014). In    spp., Tilapia spp.) and poeciliids (Gambusia spp.,
California, the invasion history of introduced fishes     Poecilia spp., Xiphophorus spp.). In Australia,
has an inherent multivariate dimensionality, with         most of the earliest (pre-1970) introductions
prior invasion success – defined as the number            of fishes that became established were species
of countries where a species was introduced and           introduced intentionally for purposes such as
became established – indicative of successful             angling and biocontrol, whereas more recent
establishment (Marchetti et al. 2004a, b, Moyle &         introductions and establishments (post-1970) were
Marchetti 2006). On the Iberian Peninsula, prior          transported via the ornamental trade (García-
invasion success of introduced fishes was a good          Díaz et al. 2018). Introductions and establishment
indicator of establishment success as well as other       of fishes in Hawaii and Guam appears to show
stages of the invasion process (Alcaraz et al. 2005,      the same trend.
Ribeiro et al. 2008). In a study of 1,678 introductions
of 280 freshwater fish species globally, the number       Many freshwater fish species present in Hawaii
of countries where a species was introduced was           and Guam were introduced via multiple pathways
a significant estimator of establishment success,         and often repeatedly, a pattern similar to fish
and all species (n = 46) introduced to nine or more       introductions on other oceanic islands. This is
countries had an established population in at least       the case with tilapia (Oreochromis spp. and Tilapia
one country (Bomford et al. 2010).                        spp.) that were repeatedly and widely distributed
J. Vertebr. Biol. 2021, 70(4): 21064                    15               Inland fish introductions on tropical Pacific islands

among Pacific islands to establish aquaculture          countries of the world (Welcomme 1981, 1988,
stocks for food, biological control agents, for use     1992). Of 32 ornamental species that dominated the
as live bait (e.g. tuna fishery), as prey for native    trade in the United States by numbers imported
species, to enhance fisheries, and for research         and monetary value (Chapman et al. 1997; Table
purposes (Gillett 1990, Nelson & Eldredge 1991,         3), 14 (44%) of the species in Table S1, or a closely
Szyper et al. 2000, De Silva et al. 2004, Nandlal       related congener, were introduced to Hawaii and/
& Pickering 2004). However, tracking the precise        or Guam, of which all but two became established
pattern of introduction and subsequent dispersal        (Pterophyllum sp. and Osteoglossum bicirrhosum).
of the different tilapia taxa is problematic due to     Moreover, there are large numbers of fish and
the existence of numerous hybrids and incorrect         other aquatic species in the ornamental fish trade
identifications. Molecular studies of both captive      that are regularly introduced on a spatial and
and naturalized populations indicate that at least      temporal basis globally to areas where they have
seven distinct tilapia lineages currently exist in      not previously been recorded (Mangiante et al.
Hawaii, including both Nile tilapia Oreochromis         2018, Chan et al. 2019), suggesting that Pacific
niloticus and O. mossambicus, with wild populations     Islands, like other freshwater regions of the world
exhibiting disproportionate hybridization and           with suitable climate, will continue to experience
introgression among species compared to captive         establishment of non-native aquarium fishes in
stocks (Szyper et al. 2000, Wu & Yang 2012). In         the future.
recent decades, the Nile tilapia has replaced
Mozambique tilapia as the most common and               Unlike previous studies (see Table 2), our a priori
widely distributed tilapia in aquaculture, partly       prediction that life-history attributes such as body
due to its rapid growth and feed efficiency (De         size, reproductive mode, or trophic category might
Silva 2012). Like other tilapia species and many        influence establishment success of fishes introduced
other cultured fish taxa, characteristics that make     to Hawaii and Guam was not supported by our
Nile tilapia attractive for aquaculture contribute to   modelling approach. There are different possible
its tendency to be highly invasive.                     reasons for this. Our data set includes a relatively
                                                        limited number of species but with a broad range of
Numerous freshwater fishes found on Pacific             life-history attributes, such that similarities in traits
Islands are associated with the ornamental garden       did not weigh proportionally as in other factors
and aquarium trade and we presume that their            that ranked high in our best models, family and
presence in the wild is primarily due to releases       prior establishment history (Table 4). Additionally,
by pet owners or escapement from ponds (Table           there could be multicollinearity between some of
S1). Many ornamental species exhibit high               our variables. Interspecific comparisons can be
propagule pressure, and most species popular            compromised by phylogenetic nonindependence,
in the aquarium trade are tropical and relatively       and species-specific variables such as body size
cold intolerant (Courtenay & Stauffer 1990, Padilla     or diet can be linked erroneously to invasion risk
& Williams 2004, Rixon et al. 2005, Duggan et           among taxonomically related species that are
al. 2006, Gertzen et al. 2008, Strecker et al. 2011).   invasive for other reasons (Ruesink 2005). Colautti
Given that Hawaii, Guam, and circum-equatorial          et al. (2006) noted that many introductions of
Pacific islands have a climate that matches regions     non-native organisms are non-random, and
where most aquarium species are native, it is           that interpretation of patterns associated with
not surprising that insular areas of Oceania and        invasiveness may be confounded by “propagule
other tropical areas of the Pacific with substantial    bias”. Our data set clearly includes examples of
human habitation are especially vulnerable to           such non-random introductions, especially groups
establishment by ornamental freshwater fishes           of fishes (e.g. centrarchids, cichlids, cyprinids,
that primarily originate from tropical and              poecilids) exhibiting taxonomic relatedness,
subtropical continental regions of Asia, Africa, and    conserved life-history features as a result of
South America. Further evidence of the substantial      phylogeny (family membership), and taxa where
propagule pressure represented by many of the           propagule pressure was particularly high.
popular aquarium taxa represented in our data
set is their disproportionately high representation     Modelling results that support high establishment
among ornamental freshwater species introduced          probability for air-breathing species (hypoxia
into the United States (Chapman et al. 1997), as well   score 3) conformed to our prediction. However,
as widespread introduction throughout numerous          our model estimate of a higher establishment
You can also read